In the US and worldwide, the vast majority of studies of alcohol's harms have focused on outcomes experienced by the drinker, given that alcohol use causes significant morbidity and mortality. However, alcohol adversely affects not only individual drinkers, but also other people, including family members, friends, co- workers and strangers. With the exception of a few areas (such as drink driving and fetal alcohol effects), alcohol's harms to others (or externalities), are largely understudied, particularly in the US. Initial studies in Australia and New Zealand suggest excessive drinking can harm families, cause child abuse or neglect, and cost people other than the drinker time, money, and peace of mind, potentially doubling the social costs directly associated with harms to drinkers themselves. There has been a complete absence of US population-wide data of alcohol's harms to others, excepting preliminary work for the present application that uses a brief scale in the National Alcohol Survey (NAS). Responding to PA-11-016 (Epidemiology & Prevention), we propose the first comprehensive investigation of harms due to others' heavy drinking, as experienced by US adults, studying types and sources of such harms, as well as impacts on health-related quality of life, mental health and time and financial burdens. To leverage the already-funded 2014-15 NAS (expected NH5,270) we plan to add 10 minutes of detailed questions for respondents reporting harm from others' drinking (expected nH1,580), as well as to field a parallel, specialized US-wide Alcohol's Harms to Others survey (NH2,250). Both surveys will include common items and scales and use the same sampling design (dual-frame landline/cell phone RDD CATI surveys, oversampling African Americans and Hispanics). This will permit pooling data for certain analyses to achieve greater power and precision. Current aims include to (1) comprehensively estimate the prevalence, sources and seriousness of alcohol's harms to others, and test hypotheses about individual differences in these; (2) test hypotheses regarding environmental influences (drinking contexts and neighborhood context, e.g., urbanicity, socioeconomic disadvantage, disorder and alcohol outlet density) to gain insight about how such harms might be prevented; (3) estimate multi-level models to examine the roles of state alcohol taxes, policies on alcohol availability, and 'wet' to 'dry' drinking cultures. We will investigate the interplay of these levels of influence on the occurrence, sources and severity of the harms. In addition to the novel direction of the research (the first US adult population survey of alcohol's harms to others), an innovation is the proposed use of propensity score matching techniques to contrast outcomes (such as mental health impacts) while addressing the place-based sorting and selection effects common in contextual analysis. As part of this work, we will identify high-risk situations associated with severe harms from others' drinking that may be amenable to preventive interventions, and examine pathways of influence, with a goal of helping to inform prevention strategies and evaluate alcohol policies that may help reduce harms from other drinkers.